COVER TYPE ANALYSIS¶
SharpestMinds Project by Angela Teng with the mentorship of Jeremie Harris
Completed:
- Loading data + importing libraries
- Exploratory Data Analysis
To Do:
- Data Cleaning
- Feature Selection
- Dimensionality Reduction
- Modeling
- Analysis
- Polish
Importing Libraries + Loading Data¶
In [50]:
import numpy as np
import scipy as sp
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from scipy.stats import norm
from sklearn.preprocessing import StandardScaler
from scipy import stats
import warnings
warnings.filterwarnings('ignore')
%matplotlib inline
from plotly import __version__
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
import cufflinks as cf
from pprint import pprint
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from sklearn.feature_selection import RFE
from sklearn.linear_model import LogisticRegression
from sklearn import metrics
from sklearn.ensemble import ExtraTreesClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import classification_report,confusion_matrix
#import sklearn.cross_validation as cross_validation
#from sklearn.cross_validation import train_test_split
import sklearn.model_selection as model_selection
from sklearn import linear_model
from IPython.display import Image
from sklearn.externals.six import StringIO
from sklearn.tree import export_graphviz
import pydot
import mca
from sklearn.neighbors import KNeighborsClassifier
from random import sample
In [2]:
import os
os.getcwd()
Out[2]:
In [3]:
os.chdir('/Users/angelateng/Documents/GitHub/Projects/Covertype_Prediction/Data')
os.getcwd()
Out[3]:
In [4]:
data = open("covtype.data")
data
Out[4]:
In [5]:
data = pd.read_csv("covtype.data", header=None)
#interesting that read csv also works on .data files!
In [6]:
data.head()
Out[6]:
In [7]:
# set column names
cols = ['elevation', 'aspect', 'slope', 'horizontal_distance_to_hydrology',
'vertical_distance_to_hydrology', 'Horizontal_Distance_To_Roadways',
'Hillshade_9am', 'Hillshade_Noon', 'Hillshade_3pm', 'Horizontal_Distance_To_Fire_Points',
'Wilderness_Area_1', 'Wilderness_Area_2', 'Wilderness_Area_3', 'Wilderness_Area_4',
'Soil_Type_1',
'Soil_Type_2',
'Soil_Type_3',
'Soil_Type_4',
'Soil_Type_5',
'Soil_Type_6',
'Soil_Type_7',
'Soil_Type_8',
'Soil_Type_9',
'Soil_Type_10',
'Soil_Type_11',
'Soil_Type_12',
'Soil_Type_13',
'Soil_Type_14',
'Soil_Type_15',
'Soil_Type_16',
'Soil_Type_17',
'Soil_Type_18',
'Soil_Type_19',
'Soil_Type_20',
'Soil_Type_21',
'Soil_Type_22',
'Soil_Type_23',
'Soil_Type_24',
'Soil_Type_25',
'Soil_Type_26',
'Soil_Type_27',
'Soil_Type_28',
'Soil_Type_29',
'Soil_Type_30',
'Soil_Type_31',
'Soil_Type_32',
'Soil_Type_33',
'Soil_Type_34',
'Soil_Type_35',
'Soil_Type_36',
'Soil_Type_37',
'Soil_Type_38',
'Soil_Type_39',
'Soil_Type_40',
'Cover_Type']
#cols
In [8]:
data.columns = cols
data.head()
Out[8]:
Exploratory Data Analysis¶
In [9]:
# check data types
data.dtypes
Out[9]:
In [10]:
data.info()
# no null values
Goal: predict cover type given the other variables (7 cover types )
In [11]:
data['Cover_Type'].describe()
Out[11]:
In [12]:
#sns.distplot(data['Cover_Type'])
#not sure histogram is what we want, kind of just want a bar chart
In [12]:
sns.set(style="whitegrid")
#ax = sns.barplot(x="Cover_Type", y ="Count", data=data['Cover_Type'])
sns.countplot(data['Cover_Type'])
Out[12]:
In [14]:
sns.distplot(data['elevation'])
#measured in meters
Out[14]:
In [15]:
sns.distplot(data['aspect'])
#measured in azimuth
Out[15]:
In [16]:
sns.distplot(data['slope'])
#measured in degrees
Out[16]:
In [17]:
sns.distplot(data['horizontal_distance_to_hydrology'])
#meters
Out[17]:
In [18]:
sns.distplot(data['vertical_distance_to_hydrology'])
#meters
#what is a negative vertical distance?
#whyy are sns charts so small
Out[18]:
In [19]:
sns.distplot(data['Horizontal_Distance_To_Roadways'])
#meters
Out[19]:
In [20]:
sns.distplot(data['Hillshade_9am'])
#index from 0-255
Out[20]:
In [21]:
sns.distplot(data['Hillshade_Noon'])
Out[21]:
In [22]:
sns.distplot(data['Hillshade_3pm'])
Out[22]:
In [23]:
sns.distplot(data['Horizontal_Distance_To_Fire_Points'])
#meters
Out[23]:
In [24]:
sns.countplot(data['Wilderness_Area_1'])
Out[24]:
In [25]:
sns.countplot(data['Wilderness_Area_2'])
Out[25]:
In [26]:
sns.countplot(data['Wilderness_Area_3'])
Out[26]:
In [27]:
sns.countplot(data['Wilderness_Area_4'])
Out[27]:
In [11]:
#plotting multiple binary categorical variables?
data.groupby('Cover_Type').describe()
Out[11]:
In [19]:
data.describe()
Out[19]:
In [30]:
#sns.pairplot(data)
#this won't run, too big
In [31]:
#correlation matrix
corrmat = data.corr()
f, ax = plt.subplots(figsize=(12, 9))
sns.heatmap(corrmat, vmax=.8, square=True);
#elevation seems important
In [32]:
#numbered/labeled corr matrix, to see var importance
k = 10 #number of variables for heatmap
cols = corrmat.nlargest(k, 'Cover_Type')['Cover_Type'].index
cm = np.corrcoef(data[cols].values.T)
sns.set(font_scale=1.25)
hm = sns.heatmap(cm, cbar=True, annot=True, square=True, fmt='.2f', annot_kws={'size': 10}, yticklabels=cols.values, xticklabels=cols.values)
plt.show()
In [33]:
#scatterplot between cover type, slope and elevation
sns.set()
cols2 = ['Cover_Type', 'slope', 'elevation']
sns.pairplot(data[cols2], size = 2.5)
plt.show();
#not 100% sure how to read this to be honest
In [34]:
#based on this plot and the corr matrix, the vars that seem most correlated with covertype are aspect, slope, and elevation
In [12]:
plt.style.use('ggplot')
In [36]:
data['Cover_Type'].plot.hist(alpha=0.5,bins=25)
Out[36]:
In [37]:
#data[data['Cover_Type']==1].plot.box()
In [42]:
sns.clustermap(data.corr(),annot=True)
Out[42]:
In [40]:
#sns.clustermap(data, metric="correlation")
In [39]:
sns.lmplot(x='Cover_Type',y='elevation',data=data)
Out[39]:
In [ ]:
#sns.pairplot(data,hue='Cover_Type',palette='rainbow')
#data set too big, doesn't run
In [21]:
#average elevation per cover type
avg_elevation = data["elevation"].mean()
#avg_elevation
#sns.boxplot(x=avg_elevation, y="Cover_Type", data=data,palette='rainbow')
data["elevation"].groupby(data['Cover_Type']).mean()
Out[21]:
In [22]:
#Average slope per covertype
data["slope"].groupby(data['Cover_Type']).mean()
Out[22]:
In [23]:
#average aspect per covertype
data["aspect"].groupby(data['Cover_Type']).mean()
Out[23]:
In [24]:
#horizontal distance to hydrology
data["horizontal_distance_to_hydrology"].groupby(data['Cover_Type']).mean()
Out[24]:
In [26]:
#vertical_distance_to_hydrology
data["vertical_distance_to_hydrology"].groupby(data['Cover_Type']).mean()
Out[26]:
In [ ]:
#sns.boxplot(x="elevation", y="Cover_Type", data=data,palette='rainbow')
#wow none of the viz work.
In [ ]:
#data.iplot(kind='scatter',x='elevation',y='Cover_Type',mode='markers',size=10)
#also doesn't run....
In [11]:
#data.iplot(kind='box')
In [17]:
cov_dummy = pd.get_dummies(data['Cover_Type'])
cov_dummy.head()
Out[17]:
In [18]:
df4 = pd.concat([cov_dummy, data], axis = 1)
df4.head()
Out[18]:
In [26]:
sns.violinplot(data = data, x='Cover_Type', y='elevation')
#sns.swarmplot(data = ds_cat, x='Cover_Type', y='elevation', color = 'k', alpha = 0.6)
Out[26]:
In [28]:
#fig = plt.figure()
#ax1 = fig.add_subplot(2,1,1)
#sns.countplot(data = data, x = 'Cover_Type', ax = ax1)
#ax2 = fig.add_subplot(2,1,2)
#sns.boxplot(data = ds_cat, x='Cover_Type', y='elevation' , ax = ax2)
In [30]:
#data.hist(figsize=(16, 20), bins=50, xlabelsize=8, ylabelsize=8)
In [31]:
cov_type_corr = data.corr()['Cover_Type'][:-1]
In [32]:
cov_type_corr
Out[32]:
In [33]:
golden_features_list = cov_type_corr[abs(cov_type_corr) > 0.5].sort_values(ascending=False)
print("There is {} strongly correlated values with Cover Type:\n{}".format(len(golden_features_list), golden_features_list))
In [34]:
corr = data.drop('Cover_Type', axis=1).corr()
plt.figure(figsize=(12, 10))
sns.heatmap(corr[(corr >= 0.5) | (corr <= -0.4)],
cmap='viridis', vmax=1.0, vmin=-1.0, linewidths=0.1,
annot=True, annot_kws={"size": 8}, square=True);
In [12]:
# can probably add additional stuff here later on.....
Feature Importance¶
Recursive Feature Elimination¶
In [13]:
# create a base classifier used to evaluate a subset of attributes
model_1 = LogisticRegression()
In [14]:
# set x and y columns
X = data.loc[:, data.columns != 'Cover_Type']
X.head()
Out[14]:
In [15]:
y = data.loc[:, data.columns == 'Cover_Type']
y.head()
Out[15]:
In [39]:
# create RFE model and select 10 attributes
rfe = RFE(model_1,3)
#rfe = rfe.fit(dataset.data, dataset.target)
#rfe = rfe.fit(x, y)
# this is not running!! dataset may be too large!!
In [40]:
# summarize the selection of the attributes
#print(rfe.support_)
#print(rfe.ranking_)
"The Recursive Feature Elimination (RFE) method is a feature selection approach. It works by recursively removing attributes and building a model on those attributes that remain. It uses the model accuracy to identify which attributes (and combination of attributes) contribute the most to predicting the target attribute."
Feature Importance¶
In [41]:
# fit an Extra Trees model to the data
model_2 = ExtraTreesClassifier()
In [42]:
model_2.fit(x, y)
Out[42]:
In [43]:
# display the relative importance of each attribute
print(model_2.feature_importances_)
In [52]:
x_cols = cols[:54]
#x_cols
In [54]:
feat_impt_val = pd.DataFrame([model_2.feature_importances_], index=None)
feat_impt_val.columns = x_cols
#feat_impt_val.columns = cols
#ValueError: Length mismatch: Expected axis has 54 elements, new values have 55 elements
feat_impt_val
Out[54]:
It seems like elevation, aspect, and slope are the three most important features.
Training and Testing Set¶
In [63]:
# train test split
X_train = x
X_train.head()
Out[63]:
In [64]:
Y_train = y
Y_train.head()
Out[64]:
In [65]:
X_train.shape
Out[65]:
In [66]:
Y_train.shape
#sanity check that the number of samples matches
Out[66]:
In [73]:
X_train, X_test, y_train, y_test = model_selection.train_test_split(x, y, train_size=0.70, random_state=101)
In [75]:
print ("X_train: ", X_train.head())
print ("y_train: ", y_train.head())
print("X_test: ", X_test.head())
print ("y_test: ", y_test.head())
In [79]:
#linear regression - trying to see which vars are most important, even tho y is categorical.... also should be doing logisitic but how do we do this on categorical data
lm = linear_model.LinearRegression()
In [83]:
model_1 = lm.fit(X_train,y_train)
In [84]:
model_1
Out[84]:
In [85]:
predictions = lm.predict(X_train)
print(predictions)
In [86]:
#R squared of the model
lm.score(X_train,y_train)
Out[86]:
In [87]:
lm.coef_
Out[87]:
In [88]:
lm.intercept_
Out[88]:
In [29]:
# WAS DOING THE ABOVE WRONG, THIS IS HOW YOU DO IT:
#X = data.drop(data['Cover_Type'], axis=1)
#y = data['Cover_Type']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.30)
Decision Trees¶
(Since Categorical)
In [32]:
dtree = DecisionTreeClassifier()
dtree
Out[32]:
In [33]:
dtree.fit(X_train,y_train)
Out[33]:
Prediction & Evaluation¶
In [34]:
predictions = dtree.predict(X_test)
In [36]:
print(classification_report(y_test,predictions))
In [37]:
print(confusion_matrix(y_test,predictions))
#not sure how to read this
Tree Viz¶
In [43]:
features = list(data.columns[:54])
features
Out[43]:
In [44]:
dot_data = StringIO()
export_graphviz(dtree, out_file=dot_data,feature_names=features,filled=True,rounded=True)
In [ ]:
#graph = pydot.graph_from_dot_data(dot_data.getvalue())
#Image(graph[0].create_png())
Dimensionality Reduction¶
"Multiple Correspondence Analysis (MCA), which is an extension of principal component analysis when the variables to be analyzed are categorical instead of quantitative (which is the case here with your binary variables). See for instance Husson et al. (2010), or Abdi and Valentin (2007)." https://stats.stackexchange.com/questions/159705/would-pca-work-for-boolean-binary-data-types
In [ ]:
src_index = (['Cover_Type1'] * 7 + ['Expert 2'] * 9 + ['Expert 3'] * 6)
In [ ]:
In [ ]:
In [48]:
#drop cols except 3 most "important".... lol bad practice but... just trying stuff
Cluster Analysis¶
In [ ]:
## KNN
In [25]:
scaler = StandardScaler()
In [26]:
scaler.fit(data.drop('Cover_Type',axis=1))
Out[26]:
In [27]:
scaled_features = scaler.transform(data.drop('Cover_Type',axis=1))
In [37]:
df_feat = pd.DataFrame(scaled_features,columns=data.columns[:-1])
df_feat.head()
Out[37]:
In [29]:
#train test split
X_train, X_test, y_train, y_test = train_test_split(scaled_features,data['Cover_Type'],
test_size=0.30)
In [37]:
#KNN: trying to come up with a model to predict cover_type 1-7 or not. We'll start with k=1.
In [41]:
knn = KNeighborsClassifier(n_neighbors=1)
In [ ]:
#knn.fit(X_train,y_train)
#jk nothing works
In [ ]:
### Just pick the first 3 variables.... basically do what you're not suppoed to do and drop everything
In [35]:
df2 = data[["Cover_Type","elevation","aspect","slope"]]
df2.head()
Out[35]:
In [36]:
scaler.fit(df2.drop('Cover_Type',axis=1))
Out[36]:
In [38]:
scaled_features = scaler.transform(df2.drop('Cover_Type',axis=1))
In [40]:
df_feat2 = pd.DataFrame(scaled_features,columns=df2.columns[1:])
df_feat2.head()
Out[40]:
In [42]:
X_train, X_test, y_train, y_test = train_test_split(scaled_features,df2['Cover_Type'],
test_size=0.30)
In [43]:
knn.fit(X_train,y_train)
Out[43]:
In [44]:
pred = knn.predict(X_test)
In [45]:
print(confusion_matrix(y_test,pred))
In [46]:
print(classification_report(y_test,pred))
In [47]:
#pick k val
error_rate = []
# Will take some time
for i in range(1,40):
knn = KNeighborsClassifier(n_neighbors=i)
knn.fit(X_train,y_train)
pred_i = knn.predict(X_test)
error_rate.append(np.mean(pred_i != y_test))
In [49]:
plt.figure(figsize=(10,6))
plt.plot(range(1,40),error_rate,color='blue', linestyle='dashed', marker='o',
markerfacecolor='red', markersize=10)
plt.title('Error Rate vs. K Value')
plt.xlabel('K')
plt.ylabel('Error Rate')
Out[49]:
In [50]:
# NOW WITH K=20
knn = KNeighborsClassifier(n_neighbors=20)
knn.fit(X_train,y_train)
pred = knn.predict(X_test)
print('WITH K=20')
print('\n')
print(confusion_matrix(y_test,pred))
print('\n')
print(classification_report(y_test,pred))
#lol still bad
In [53]:
# D Tree with just 3 features
In [55]:
X = df2.drop('Cover_Type',axis=1)
y = df2['Cover_Type']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.30)
In [56]:
dtree = DecisionTreeClassifier()
dtree.fit(X_train,y_train)
Out[56]:
In [59]:
predictions = dtree.predict(X_test)
print(classification_report(y_test,predictions))
print(confusion_matrix(y_test,predictions))
In [61]:
features = list(df2.columns[1:])
features
Out[61]:
In [ ]:
dot_data = StringIO()
export_graphviz(dtree, out_file=dot_data,feature_names=features,filled=True,rounded=True)
graph = pydot.graph_from_dot_data(dot_data.getvalue())
Image(graph[0].create_png())
Sampling the Population¶
In [16]:
# do you sample before you normalize? and log transform? or after
In [19]:
#dataframe with dummy variables for covertype
df4.head()
Out[19]:
In [45]:
df_dummy = df4
#df_dummy.drop(['Cover_Type'], axis=1)
Out[45]:
In [47]:
df_dummy = df_dummy.drop(['Cover_Type'], axis=1)
df_dummy.head()
Out[47]:
In [48]:
df_dummy.columns
Out[48]:
In [71]:
?train_test_split
?sample
?loc
?concatenate
?pd.concat
In [67]:
#sample(df_dummy['1'], 1000)
#sample(df_dummy, 1000)
#df_dummy.sample(1000)
# select a random sample of 1000 entries per cover type
CV1 = df_dummy[df_dummy[1]==1].sample(1000)
CV2 = df_dummy[df_dummy[2]==1].sample(1000)
CV3 = df_dummy[df_dummy[3]==1].sample(1000)
CV4 = df_dummy[df_dummy[4]==1].sample(1000)
CV5 = df_dummy[df_dummy[5]==1].sample(1000)
CV6 = df_dummy[df_dummy[6]==1].sample(1000)
CV7 = df_dummy[df_dummy[7]==1].sample(1000)
CV1.head()
CV2.head()
Out[67]:
In [76]:
# concatenate those 1000 entries to get a sample size of 7000
# need to make sure to de-dupe this
sample = pd.concat((CV1, CV2, CV3, CV4, CV5, CV6, CV7), axis=0)
sample.head()
Out[76]:
In [77]:
sample.shape
Out[77]:
In [81]:
# next we want to do more exploratory analysis on this random sample.
sample.describe()
Out[81]:
In [80]:
sns.pairplot(sample)
Out[80]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
Sources:¶
Data Sources and Documentation:
- https://archive.ics.uci.edu/ml/datasets/Covertype
- http://rexa.info/paper/d44ba72ee840ac929fb2b78f6d9f6c6b408ff1da
Extraction:
- https://realpython.com/python-string-split-concatenate-join/
- https://stackoverflow.com/questions/28595701/pandas-equivalent-of-rs-cbind-concatenate-stack-vectors-vertically
- https://stackoverflow.com/questions/34091877/how-to-add-header-row-to-a-pandas-dataframe
- https://www.shanelynn.ie/python-pandas-read_csv-load-data-from-csv-files/
- https://www.guru99.com/reading-and-writing-files-in-python.html
- https://stackoverflow.com/questions/31797013/how-to-open-a-data-file-extension
Exploration:
- https://www.datacamp.com/community/tutorials/categorical-data
- https://www.kaggle.com/pmarcelino/comprehensive-data-exploration-with-python
- https://stackoverflow.com/questions/29763620/how-to-select-all-columns-except-one-column-in-pandas/29763653
- https://www.kaggle.com/pmarcelino/comprehensive-data-exploration-with-python
- https://jakevdp.github.io/PythonDataScienceHandbook/04.14-visualization-with-seaborn.html
- https://seaborn.pydata.org/generated/seaborn.countplot.html
- https://seaborn.pydata.org/tutorial/categorical.html
- Categorical Data https://www.datacamp.com/community/tutorials/categorical-data
- https://stackoverflow.com/questions/43131274/how-do-i-plot-two-countplot-graphs-side-by-side-in-seaborn
- https://stats.stackexchange.com/questions/159705/would-pca-work-for-boolean-binary-data-types
- https://www.kaggle.com/nextbigwhat/eda-for-categorical-variables-a-beginner-s-way
- https://towardsdatascience.com/dealing-with-categorical-data-fast-an-example-d4329b44253d
- https://github.com/samirgadkari/Predict_Functional_Water_Pumps_Tanzania_Kaggle/blob/master/Predictive_Modeling_Kaggle_Challenge.ipynb
- https://seaborn.pydata.org/generated/seaborn.clustermap.html
- https://stackoverflow.com/questions/42818670/pandas-dataframe-average-calculation
- https://www.kaggle.com/ekami66/detailed-exploratory-data-analysis-with-python
General Syntax:
- https://stackoverflow.com/questions/42202872/how-to-convert-list-to-row-dataframe-with-pandas
- https://chrisalbon.com/python/data_wrangling/pandas_rename_multiple_columns/
- https://stackoverflow.com/questions/42049147/convert-list-to-pandas-dataframe-column/42049158#42049158
- https://stackoverflow.com/questions/33575587/pandas-dataframe-how-to-apply-describe-to-each-group-and-add-to-new-columns
- https://towardsdatascience.com/the-dummys-guide-to-creating-dummy-variables-f21faddb1d40
- https://cmdlinetips.com/2018/04/how-to-drop-one-or-more-columns-in-pandas-dataframe/
- https://stackoverflow.com/questions/11346283/renaming-columns-in-pandas
- https://pbpython.com/categorical-encoding.html
- https://www.geeksforgeeks.org/python-random-sample-function/
Data Sampling:
- https://towardsdatascience.com/understanding-feature-engineering-part-2-categorical-data-f54324193e63
- https://github.com/dipanjanS/practical-machine-learning-with-python#contents
Feature Selection:
- https://medium.com/@aneesha/recursive-feature-elimination-with-scikit-learn-3a2cbdf23fb7
- https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFE.html#sklearn.feature_selection.RFE
- https://machinelearningmastery.com/feature-selection-in-python-with-scikit-learn/
- https://machinelearningmastery.com/feature-selection-in-python-with-scikit-learn/
- https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFE.html#sklearn.feature_selection.RFE
- https://medium.com/@aneesha/recursive-feature-elimination-with-scikit-learn-3a2cbdf23fb7
- PCA:
- https://scikit-learn.org/stable/auto_examples/neighbors/plot_nca_dim_reduction.html#sphx-glr-auto-examples-neighbors-plot-nca-dim-reduction-py
- http://staskolenikov.net/talks/Gustavo-Stas-PCA-generic.pdf
- https://www.researchgate.net/post/Should_I_use_PCA_with_categorical_data
- https://stackoverflow.com/questions/40795141/pca-for-categorical-features
- Dimensionality Reduction:
- https://towardsdatascience.com/understanding-feature-engineering-part-2-categorical-data-f54324193e63
- https://www.kaggle.com/questions-and-answers/55494
- https://datascience.stackexchange.com/questions/36849/how-to-implement-feature-selection-for-categorical-variables-especially-with-ma
Transformation:
Modeling:
- https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html
- https://www.kdnuggets.com/2016/03/data-science-process.html
- https://towardsdatascience.com/train-test-split-and-cross-validation-in-python-80b61beca4b6
- https://github.com/janeyin600/Data_analysis/blob/master/train_test_split.ipynb
- https://stackoverflow.com/questions/30667525/importerror-no-module-named-sklearn-cross-validation
- https://stats.stackexchange.com/questions/189760/two-ways-of-predicting-a-categorical-variable
- https://towardsdatascience.com/train-test-split-and-cross-validation-in-python-80b61beca4b6
- https://github.com/samirgadkari/Predict_Functional_Water_Pumps_Tanzania_Kaggle/blob/master/Predictive_Modeling_Kaggle_Challenge.ipynb
- https://towardsdatascience.com/dealing-with-categorical-data-fast-an-example-d4329b44253d
- Linear Regression:
- Softmax Regression:
Loading:
Questions:¶
- how do I use RFE??? AttributeError: 'DataFrame' object has no attribute 'data'
- deep dive into explaining/understanding a correlation matrix